Chapter 4 David Nowak

Assessing the Sustainability of Agricultural and Urban Forests in the United States

Urban Forests

Urban forests (and trees) constitute the second forest resource urban development. The characteristics of these forests are considered in this report. We specifically emphasize the fact determined by both their natural components and the anthropo­ that agricultural and urban forests exist on a continuum defined genic elements in the landscapes in which they occur. by their relationship (and interrelationship) with a given land­ scape. These two forest types generally serve different purposes, Definitions however. Whereas agricultural forests are considered primarily in terms of their contribution to biodiversity conservation or, For purposes of this report, urban forests are composed of all as in the case of agroforestry, to agricultural production, urban the trees within our urban lands. The definition conceptually forests are assessed primarily in terms of the range of environ­ extends to include the various ecosystem components that mental services and values they provide to urban and suburban accompany these trees (e.g., soils or understory flora), although residents. The potential list of services is extensive and will we do not explicitly identify all these components. Urban forests vary according to different individuals, organizations, and loca­ can contain forested stands, like in rural areas, but they also tions, with many services being difficult to precisely quantify. contain trees found along streets, in residential lots, in parks, Trees affect numerous environmental processes, such as water and in other land uses. The forests are a mix of planted and cycling; sound propagation; and pollution formation, dispersion, naturally regenerated trees. For data gathering and reporting and removal. Trees also directly affect human populations by purposes, the key to defining urban forests lies in the definition altering the social, economic, health, and aesthetic aspects of of what precisely constitutes urban land. Using the Census urban environments. These effects exist in all treed landscapes Bureau’s definition,urban land consists of all territory, popula­ but are more prominent in urban areas because of the higher tion, and housing units located within either urbanized areas or concentration of people. urban clusters (Census Bureau 2014).

As in the previous chapter, this chapter begins with a general Urbanized areas consist of densely settled territories that contain description of the resource, including formal definitions. This 50,000 or more people; urban clusters consist of densely settled first section also includes a brief listing of environmental territories that have at least 2,500 people but fewer than 50,000 services associated with urban forests and the specific threats people (fig. 4.1). Urbanized area and urban cluster boundaries they face. The second section presents currently available data encompass densely settled territories and are defined by— for understanding urban forests at the national scale. These data ƒƒ A cluster of one or more block groups or census blocks with rely heavily on satellite imagery and are focused on describing a population density of at least 1,000 people per square mile. the extent of forest cover in urban areas. The chapter concludes with a discussion of the adequacy of the current information ƒƒ Surrounding block groups and census blocks with a popula­ base and strategies for improving it. tion density of 500 people per square mile. ƒƒ Less densely settled blocks that form enclaves or indentations or that are used to connect discontinuous areas (Census Definitions and Characteristics Bureau 2014). of Urban Forest Ecosystems This definition of urban lands is based solely on census blocks Urban forests are defined by their proximity to human popula­ and their population density. Census blocks, in turn, are deter­ tions and include numerous physical elements that constitute mined in part by physical features on the land, both constructed, such as roads and rail lines, and natural, such as rivers and

Assessing the Sustainability of Agricultural and Urban Forests in the United States 37 Figure 4.1. Urban areas in the contiguous United States, 2000, based on the Census Bureau definition of urban land (Nowak and Dwyer 2007).

ridgelines. (Additional information is available at the Census the only “nature” some urbanites experience in their lives is Bureau’s Web site—http://www.census.gov/ [August 2015].) from contact with urban forests. These trees and forests provide The resulting definitions of urban lands will not always match an array of species and structural diversity that is not typically the jurisdictional boundaries of cities and towns. Urban forests, found in other forests. Species richness and diversity in urban however, are most commonly managed at the municipal level. forests are typically greater than what is found in surrounding A uniform definition that can consistently span different jurisdic­ native stands, with urban forests containing varying proportions tional boundaries is essential for developing consistent national of nonnative tree species (Nowak 2010). Not only are species and regional inventories, especially if the inventories are to diverse, but the tree configurations in urban areas also can be be combined with inventories from other land use classes, but diverse, crossing many land use types and including single tree jurisdictional boundaries will often be crucial in determining specimens, linear rows of street trees or trees along fence rows, how, and if, forest resources will be managed. Assessments and large patches of intact forest stands. The diversity of trees conducted within jurisdictional boundaries (community forests) is often dwarfed by the diversity of landowners in urban areas. and urban boundaries (urban forests) can be found within State The ownership of trees ranges from numerous small parcels of reports at http://www.nrs.fs.fed.us/data/urban (August 2015). family homes, to private commercial tracts, to varying-sized Information about wildland-urban interfaces, or WUIs, is available public properties with varying densities of trees. Urban trees at http://www.nrs.fs.fed.us/disturbance/fire/wui/ (August 2015). include a mix of planted and naturally regenerated species (Nowak 2012) and often are managed to sustain tree health and Characteristics of Urban Forests benefits and to minimize risk to or conflict with human popula­ tions. They typically are not managed as a crop to be harvested; Urban forest ecosystems have many special characteristics that, rather, they are a landscape element to be enhanced or sustained. in combination, distinguish them from other forest types. These characteristics include (1) close proximity to large or dense human populations, (2) relatively high diversity of species and Urban Forest Sustainability forest patch structures, (3) multiple public and private owner­ One main objective of urban forest management is to provide ship types, and (4) management often geared toward sustaining for optimal and sustainable benefits from trees for current and tree health and ecosystem services. More than 80 percent of the future generations. To promote optimum sustainability, managers U.S. population lives in urban areas; thus, urban forests greatly need to understand the current resource and how it is changing influence the day-to-day lives of most Americans. so they can properly guide the resource to a desired future state. Tree cover in urban areas has been declining in recent years These influences include positive and negative experiences (see (Nowak and Greenfield 2012b) and tree cover is constantly Associated Benefits and Costs in the following section). Often, changing due to various natural and anthropogenic forces.

38 Assessing the Sustainability of Agricultural and Urban Forests in the United States Natural forces for change include natural regeneration, tree Urban trees reduce carbon dioxide (CO2), a major greenhouse growth and tree mortality from insects and diseases, storms, gas, by directly removing it from the atmosphere and storing fire, old age, etc. Anthropogenic factors that influence tree (“sequestering”) the carbon in the trees as biomass. By reduc­ cover include tree planting and tree mortality or removal from ing building energy use, trees can also reduce the emission either direct or indirect human actions such as development and of CO2 from power plants. Tree-maintenance activities often pollution. The combination of these factors through time deter­ require the use of fossil fuels that emit CO2, however, and mines existing and future forest structure, species composition, improperly located trees around buildings can increase energy and tree-cover levels. demands and consequent emissions of CO2 (e.g., Nowak 2000; Nowak et al. 2002b). Sustaining desired levels of services or benefits is most easily related to sustaining a certain level and distribution of tree cover. Air quality. Trees influence air quality in a number of ways. Sustaining a desired level of canopy cover requires ensuring Trees remove pollution from the air by intercepting airborne an adequate establishment of new trees (via planting or natural particles on their leaves and branches, and absorbing gaseous regeneration) to offset loss in tree canopy due to tree mortality. pollutants into their leaves via stomata. Pollution removal by Determining the exact tree establishment rate is difficult because trees within a city can be on the order of thousands of tons trees grow (increasing canopy through time), trees are different annually, with air-quality improvement typically less than 1 sizes (canopy loss from the removal of one large tree cannot be percent (Nowak et al. 2006a). Trees also emit various volatile replaced by planting one small tree), and the system is constantly organic compounds that can contribute to the formation of changing due to human (e.g., development) and natural (e.g., ozone (O3). By lowering air temperatures via transpirational storms) factors that can create drastic cover changes in a short cooling and shading, however, trees lower the emission of period of time. Although sustaining canopy cover is important, volatile organic compounds from vegetation and numerous it is different from optimizing canopy cover, which requires ­anthropogenic sources (e.g., gasoline), thus reducing the poten­ additional information on species and locations to ensure the tial for ozone formation. In addition, trees can produce pollen optimal distribution of benefits at minimal cost over time. that can exacerbate allergies. Finally, by reducing building en­ ergy requirements, trees reduce pollutant emissions from power Monitoring urban forests is critical to ensure sustainable, opti­ plants, thereby improving air quality (Nowak 1994; Nowak et mal, and healthy urban forests. Monitoring data can be used to al. 2006a; Nowak and Dwyer 2007). detect changes and determine if management plans are meeting their desired goals. By monitoring, managers can better un­ Urban hydrology. By intercepting and retaining or slowing derstand how the resource is changing and management plans the flow of precipitation reaching the ground, urban forests can can be adjusted to ensure healthy urban forests that meet the play an important role in urban hydrologic processes. They desired goals of the local residents and sustain forest benefits can reduce the rate and volume of stormwater runoff, flooding for future generations. damage, and stormwater treatment costs, and they can enhance water quality. Estimates of runoff for an intense storm in Benefits and Costs Associated With Dayton, OH, for example, showed that the existing tree canopy Urban Forests reduced potential runoff by 7 percent; a modest increase in the canopy would have reduced runoff by nearly 12 percent Urban trees provide innumerable annual ecosystem services (Sanders 1986). The greatest percent of rainfall interception that affect the local physical environment (such as air and water occurs during the more common small storm events. During quality) and the social environment (such as individual and large rain events, the percent of rainfall interception can drop community well-being). These services can positively influence to a very small percent as most of the rain reaches the ground. urban quality of life but also have various costs (Nowak and During these large storm events, trees exert a relatively small Dwyer 2007). Urban forest services (benefits) and disservices effect from rainfall interception. To better manage storm (costs) include, but are not limited to, the following. runoff, a number of U.S. cities are moving forward with the use of enhanced tree plantings in combination with other “green Energy conservation and carbon dioxide sequestration. Trees infrastructure” in lieu of expanded pipe and culvert networks, reduce energy needs for heating or cooling buildings by shad­ or “grey infrastructure” (Philadelphia’s Green Infrastructure ing buildings in the summer, reducing summer air temperatures Plan is a notable example). (primarily through transpirational cooling), and by blocking winter winds. Trees also can increase heating needs, however, Noise reduction. Properly designed plantings of trees and by shading buildings in the winter if planted in improper loca­ shrubs can significantly reduce noise levels (Anderson et al. tions close to structures. The energy effects of trees vary with 1984). Wide belts (approximately 100 feet [30 meters]) of tall regional climate and their location around the building (Heisler dense trees combined with soft ground surfaces can reduce 1986).

Assessing the Sustainability of Agricultural and Urban Forests in the United States 39 apparent loudness by 50 percent or more (6 to 10 decibels) limbs can be used to produce various wood products or fuels (Cook 1978). Although noise reduction from plantings along (e.g., fire wood or ethanol) that can be used by residents, while roadsides in urbanized areas often is limited due to narrow creating additional jobs in the process. roadside planting space (less than 10 feet [3 meters] in width), Management costs. Although natural regeneration is a powerful reductions in noise of 3 to 5 decibels can be achieved with nar­ force in shaping the urban forest (Nowak 2012), tree planting row dense vegetation belts with one row of shrubs roadside and and various maintenance activities (e.g., watering, raking, one row of trees behind (Reethof and McDaniel 1978). pruning, tree removals) incur economic costs while helping Quality of life. The presence of urban trees can make the to provide for safe and healthy urban forests. Enhancing tree urban environment a more aesthetic, pleasant, and emotionally cover in environments that tend to be precipitation limited satisfying place in which to live, work, and spend leisure time involves additional economic and environmental costs. Planting (Dwyer et al. 1991; Taylor et al. 2001a, 2001b; Ulrich 1984). trees in these environments can produce substantial benefits Studies of urbanites’ preferences and behavior have confirmed for the urban population, but such plantings often require the strong contribution of trees and forests to the quality of life water or economic resources that may be scarce. In addition to in urban areas. Urban forests also provide significant outdoor management costs, various risks associated with urban forests leisure and recreation opportunities for urbanites (e.g., Dwyer related to falling trees and limbs may pose additional costs 1991, Dwyer et al. 1989). through personal injury, property damage, and power outages. Proper management and maintenance can minimize risks and Urban forest environments provide aesthetically pleasing sur­ costs, while enhancing numerous benefits for current and future roundings, increased enjoyment of everyday life, and a greater generations. Disposal of leaves and other detritus can incur sense of connection between people and the natural environ­ significant cost but also represents a potentially valuable supply ment. Trees are among the most important features that contrib­ of wood or organic matter (e.g., for mulch, wood products or ute to the aesthetic quality of residential streets and community bioenergy applications). parks (Schroeder 1989). Perceptions, such as aesthetic quality and personal safety, are highly sensitive to features of the urban forest such as number of trees per acre and viewing distance Major Threats and Influences Affecting (Schroeder and Anderson 1984). the Urban Forest

Community well-being. Urban forests make important con­ Numerous potential threats can significantly alter urban forests tributions to the vitality and character of a city, neighborhood, and their associated benefits. These threats (Nowak et al. 2010) or subdivision. Furthermore, the act of planting and caring for include the following. trees, when undertaken by residents, yields important social Insects and diseases. Urban forests can be, and are, severely benefits and a stronger sense of community. In addition, em­ affected by numerous insects and diseases, many of them intro­ powerment to improve neighborhood conditions in inner cities duced from other geographic regions into urban centers. Some has been attributed to involvement in urban forestry efforts insects and diseases—such as the gypsy moth, Asian longhorned (Kuo and Sullivan 2001a, 2001b; Sommer et al. 1994a, 1994b; beetle, emerald ash borer, and Dutch elm disease—have caused Westphal 1999, 2003). significant tree mortality that has virtually eliminated dominant Physical and mental health. Reduced stress and improved tree species in some places (e.g., Dozier 2012, Liebhold et al. physical health for urban residents have been associated with 1995). the presence of urban trees and forests. Landscapes with trees Wildfire. Uncontrolled fires can cause significant damage to and other vegetation have produced more relaxed physiological trees and forests and dramatically alter the urban landscape, states in humans than landscapes without these natural features. especially in urban areas adjacent to wildlands (Nowak 1993, Hospital patients with window views of trees recovered sig­ Spyratos et al. 2007). High population growth and urban expan­ nificantly faster and with fewer complications than comparable sion in California, for example, have led to a substantial increase patients without such views (Ulrich 1984). in fire ignitions in wildland-urban interface areas (Syphard et Local economic development. Urban forest resources con­ al. 2007). In addition, the intermingling of trees with manu­ tribute to the economic vitality of a city, neighborhood, or factured structures in these areas significantly complicates and subdivision. By improving the environment, trees contribute to limits the options available for fire suppression activities and increased property values, sales by businesses, and employment vegetation management practices used to reduce fire risk. (e.g., Anderson and Cordell 1988; Corrill et al. 1978; Donovan Storms. Urban forests can be altered and have been significantly and Butry 2008; Dwyer et al. 1992; Wolf 2003, 2004). Urban damaged by wind, ice, and snow storms that result in broken forest maintenance and management activities also create jobs branches and toppled trees (e.g., Greenberg and McNab 1998, to help the local economy, and wood from removed trees and Irland 2000, Proulx and Greene 2001, Valinger and Fridman

40 Assessing the Sustainability of Agricultural and Urban Forests in the United States 1997). As in the case of fire, the proximity of trees to buildings, Improper management. Because numerous people directly roads, and power lines complicates forest management in this manage most of the urban forest, the decisions and actions of regard, while elevating the potential damage that can result. the managers significantly affect urban forest composition and health. Improper decisions related to species selection, tree Invasive plants. Invasive plants such as kudzu (Pueraria locations, and maintenance can lead to conflicts with the urban lobata), English ivy (Hederal helix), European buckthorn population and infrastructure, tree damage, and poor tree health (Rhamnus cathartica), and Norway maple (Acer plantanoides) that can lead to premature tree mortality. Actions or inactions can degrade or alter urban forests by removing and replacing taken by the multitude of urban landowners can pose a threat native plants and altering ecosystem structure. English ivy and to urban forests, but they can also help bolster urban forest kudzu have been known to cover acres of canopy trees (Dozier health and sustainability if proper tree care and management 2012, Webb et al. 2001). The introduction of nonnative species are conducted. in gardens and parks enhances this risk.

Development. Land development significantly alters the urban landscape, affecting plant and wildlife populations and forest Currently Available Data on U.S. biodiversity and health (Nowak et al. 2005). Development can Urban Forests lead to rapid reductions in tree populations (clearing of forest stands), can alter species composition (e.g., tree planting after Although far from complete for national assessment needs, the development), can increase tree populations (e.g., tree planting data describing urban forests in the United States is improving. in formerly cleared areas), and can alter the urban environment Remote-sensing techniques are being used to construct urban (e.g., increase or decrease in air temperatures). Development forest cover estimates at the local and regional level across the associated with urban expansion into rural areas can also sig­ country. In addition, new urban forest field data are continually nificantly alter the regional landscape, particularly in forested being collected by local groups and cities, and through the Forest­ regions where forest area is reduced, fragmented, or parcelized Inventory and Analysis (FIA) program of the U.S. Department (i.e., forest stands remain intact but have multiple landowners). of Agriculture, Forest Service in select metropolitan areas. In timber-producing regions, when development alters the rural Tools are being designed and improved to help municipalities forest landscape, it will likewise affect the available timber inventory their forests in a consistent fashion while fostering supply and forest management practices (Zhang et al. 2005). the participation of interested citizens.

Pollution. Air and water pollution can affect tree health in urban areas if pollutant concentrations reach damaging levels. Extent of Urban Land in the United Forests have been shown to be affected by air pollution, States especially from regional deposition of ozone, nitrogen, sulfur, The importance of urban forests and their benefits in the United and hydrogen (Stolte 1996). Ozone has been documented to States is increasing because of the expansion of urban land. The reduce tree growth (Pye 1988), reduce resistance to bark beetle, percent of the coterminous United States classified as urban and increase susceptibility to drought (Stolte 1996). Air pollu­ increased from 2.5 percent in 1990 to 3.1 percent in 2000, an tion can also enhance tree growth through increased levels of area about the size of Vermont and New Hampshire combined. carbon dioxide or by providing essential plant nutrients such as The States with the highest percent urban land are New Jersey sulfur and nitrogen (e.g., NAPAP 1991). (36.2 percent), Rhode Island (35.9 percent), Connecticut (35.5 percent), and Massachusetts (34.2 percent), and 7 of the top 10 Climate change. Climate change is expected to produce most urbanized States are located in the Northeastern United warmer air temperatures, altered precipitation patterns, and States (fig. 4.2). Urban land in the coterminous United States in more extreme temperature and precipitation events (EPA 2009, 2010 increased to 3.6 percent (U.S. Census 2014). Most of the IPCC 2007). These climate changes can cause changes in 1990-to-2000 urban expansion occurred in previously forested urban forest composition (Iverson and Prasad 2001, Johnston areas (33.4 percent of the expansion) or agricultural lands 2004) and have the potential to exacerbate other urban forest (32.7 percent). The dominant type of land uses or cover classes threats (e.g., invasive species and pests). Climate change has occurring in a given State largely determines the type of land the potential to alter urban forests, not only through species being converted to development. States where more than 60 changes, but also through direct effects from storms, floods, percent of urban land expansion occurred in forests were Rhode etc., that may kill large portions of the forest in relative short Island (64.8 percent of urban expansion), Connecticut (64.1 time periods. Urban forest managers will need to understand per­cent), Georgia (64.0 percent), Massachusetts (62.9 percent),­ and adapt to potential species shifts and changes to the environ­ West Virginia (62.2 percent), and New Hampshire (61.3 percent). ment to produce sustainable and healthy urban forests under States where more than 60 percent of urban land expansion future climatic conditions.

Assessing the Sustainability of Agricultural and Urban Forests in the United States 41 Figure 4.2. Percent of U.S. counties classified as urban, 2000 (Nowak et al. 2010).

occurred in agricultural lands were Nebraska (68.9 percent), percent), New Jersey (5.1 percent), Connecticut (5.0 percent), Indiana (66.8 percent), Illinois (64.8 percent), and Wisconsin Massachusetts (5.0 percent), Delaware (4.1 percent), Maryland (62.0 percent) (Nowak et al. 2005). These estimates of urban (3.0 percent), and Florida (2.5 percent). States with the greatest land and urban land expansion within land cover types are based absolute increase in urban land are Florida (925,000 acres; on Census Bureau maps of urban land and National Land Cover 374,000 hectares), Texas (871,000 acres; 352,000 hectares) and Database (NLCD) maps of land cover types. Although these California (737,000 acres; 298,000 hectares) (Nowak et al. 2005). maps may have some inaccuracies, the urban land maps are Given the urban growth patterns of the 1990s, urban land is fairly accurate because they are based on extensive census data. projected to expand from 3.1 percent of conterminous United The most urbanized regions of the United States are the North­ States in 2000, to 8.1 percent in 2050, an increase in area east (9.7 percent of total land area) and Southeast (7.5 percent), greater than the size of Montana (fig. 4.3). The total projected with these regions also exhibiting the greatest increase in amount of U.S. forest land projected to be subsumed by urban­ percent urban land between 1990 and 2000 (1.5 and 1.8 per­ ization between 2000 and 2050 is about 29.2 million acres (11.8 cent, respectively). States with the greatest increase in percent million hectares), an area approximately the size of Pennsylva­ urban land between 1990 and 2000 were Rhode Island (5.7 nia (Nowak and Walton 2005) (fig. 4.4).

42 Assessing the Sustainability of Agricultural and Urban Forests in the United States Figure 4.3. Percent of land classified as urban in 2000 (a) and projected percent of land classified as urban in 2050 (b), by county (Nowak and Walton 2005).

(a)

(b)

Assessing the Sustainability of Agricultural and Urban Forests in the United States 43 Figure 4.4. Percent (a) and square kilometers (b) of nonurban forest subsumed by projected urban growth, 2000 through 2050, by State (Nowak and Walton 2005).

(a)

(b)

44 Assessing the Sustainability of Agricultural and Urban Forests in the United States Urban Tree-Cover Estimates metric tons) of carbon (Nowak et al. 2013), which is valued at $94.3 billion, or $146.7 per metric ton of carbon (Interagency Tree cover in urban lands in the United States (circa [ca.] 2005) Working Group 2015, U.S. EPA 2015). Soils in urban areas is currently estimated at 35.0 percent (Nowak and Greenfield also store about 2.1 billion tons of carbon in the United States 2012a). Urban tree cover has declined slightly in recent years (Pouyat et al. 2006). These values are current asset values that (ca. 2002 to 2009) with a loss of about 4 million urban trees per would be lost through the loss of existing urban forests nationally. year (Nowak and Greenfield 2012b). These estimates are based on photo interpretation of tree cover nationally. Urban tree cov­ In addition to these structural values, urban forests annually er varies across the United States, with urban tree cover tending provide substantial functional values or benefits based on the to be highest in forested regions, followed by grasslands and ecosystem services they provide. Some of these functional ben­ deserts (Nowak et al. 2001). In addition to photo-interpretation efits accrue in the tree (e.g., carbon storage) and can be partially estimates, tree-cover maps for the United States have been or completely lost when the trees die (e.g., tree decomposition produced based on 30-meter resolution satellite data (ca. 2001) can release carbon back to the atmosphere). Other functional as part of the NLCD (USGS 2008) (fig. 4.5). These tree-cover values do not accrue within a tree, however, and thus are con­ maps underestimate tree cover on average by 9.7 percent, with tinuously gained as long as the forest is healthy and functioning underestimation varying across the conterminous United States and will not be lost when the forest dies (e.g., accrued energy (Greenfield et al. 2009, Nowak and Greenfield 2010). savings from temperature moderation and shade during a tree’s lifetime are not lost when the tree dies). Based on field data sampled from several cities combined with national urban tree-cover estimates, an estimated 3.8 billion Urban trees in the conterminous United States remove ap­ urban trees are growing in the conterminous United States (the proximately 651,000 metric tons (717,000 tons) of air pollution actual range of the national estimate is between 2.4 and 5.7 billion annually, with a value of $4.7 billion (Nowak et al. 2014). trees, based on minimum and maximum city tree-cover density These trees also annually sequester about 25.6 million metric estimates) (Nowak et al. 2001). These trees have an estimated tons of carbon (28.2 million tons of carbon) (Nowak et al. structural asset value of $2.4 trillion (Nowak et al. 2002a). The 2013), or $3.8 billion per year based on 2015 estimates of the structural value averages about $600 per tree and is based on total cost of a unit of carbon emissions to a society (Interagency a formula from the Council of Tree and Landscape Appraisers Working Group 2015, U.S. EPA 2015). In addition, the total that estimates the replacement or compensatory (if a tree is too annual contribution of trees in urban parks and recreation areas large to be directly replaced) value of a tree. Structural values to the value of recreation experiences provided in the United vary by location, tree size, species, and condition of the tree. The States could exceed $2 billion (Dwyer 1991). urban forest nationally stores about 709 million tons (643 million

Figure 4.5. Percent tree cover in urban areas, 2000, by county (Nowak et al. 2010).

Assessing the Sustainability of Agricultural and Urban Forests in the United States 45 Local Inventory Activities and Related general, are piecemeal in nature and are not always consistent Tools with efforts undertaken elsewhere. Alone, these local inven­ tories cannot be readily used as building blocks for a national The foregoing text has stressed information available at the assessment. The State urban forest assessments are geared to national scale, relying primarily on remote-sensing data of providing consistent urban forest data at the State scale, but urban tree cover combined with various one-time analyses of they are currently of limited usefulness at the national scale due city tree populations to derive estimates of national urban tree to the small number of State assessments so far completed. totals and economic values. Urban forests, however, are man­ aged at the local level and, in recent years, local citizens and The planned implementation of an FIA urban forest inventory interest groups have become increasingly engaged in assessing and monitoring program at the national level will fill a major urban forest resources and improving their condition through information gap in the effort to improve urban natural resource management. New tools have emerged to facilitate this activity. stewardship (Cumming et al. 2008). The starting of FIA meas­ The i-Tree Eco model is one example, in which local residents urement and monitoring in Baltimore, Austin, and other cities are encouraged to use standard sampling protocols in their local will facilitate better linking and consistency among FIA urban area to ensure the collection of consistent data that measure and conventional forest inventories and also inclusion within important aspects of urban forests (Nowak et al. 2008).19 Since i-Tree. This new national FIA urban program is expanding to i-Tree’s introduction in 2006, there have been more than 60,000 other cities and will provide more useful data at the local scale users in more than 120 countries, with user downloads increas­ due to increased sample sizes within the cities. These data will ing at a rate of about 25 percent per year. This growth reflects provide critical baseline information and monitoring data from the desire by citizens and managers to better understand the eco­ local to regional scales. These local-scale analyses will provide system services that urban forests provide. The i-Tree provides limited information, however, for a national assessment. Until a foundation for a growing database on local forest conditions, all metropolitan areas are assessed, development of a national but it does not constitute a consistent national survey of urban urban forest assessment will be challenging. forest resources in the United States. Local-scale urban forest information can be, and is being, ana­- lyzed using i-Tree, but these data are collected by various groups Building a National Inventory with varying degrees of quality control and are not an adequate In an effort to better understand the urban forest resource at the substitute for field data gathered through a national inventory in national scale, urban forest inventory methods were pilot-tested a consistent fashion across space and time. This critical infor­ in five States (Colorado, Indiana, New Jersey, Tennessee, and mation gap needs to be filled to fully assess and understand our Wisconsin; Cumming et al. 2008). Statewide urban forest Nation’s urban forest resources. Information from a national inventories have also been conducted more recently in Alaska, survey can be used to better understand the magnitude of this California, Hawaii, Oregon, and Washington. Data from these resource, and how it is changing through time, so that better State assessments are run through the i-Tree model to provide management plans and policies can be developed to sustain estimates of ecosystem services and values. This activity is the and enhance urban forest benefits for future generations. This foundation for the full implementation of a national urban forest understanding, in turn, will enable us to disseminate improved inventory and monitoring program of major metropolitan areas best practices, identify emergent threats, and devise national that started in 2014 with the FIA program staff measuring field and regional policies and partnerships aimed at improving plots in Baltimore, MD, and Austin, TX. stewardship of these valuable resources. If integrated with con­ ventional forest inventory activity through the FIA program, along with similar surveys of agricultural forests resources, a Assessing Data Adequacy and national urban forest inventory would constitute an essential piece of the information base needed to successfully engage Identifying Major Data Gaps in landscape-scale resource conservation that bridges jurisdic­ Although i-Tree Eco assessments and urban forest assessments tional boundaries, ownership classes, and land use types. by FIA are increasing, major gaps still exist in basic urban for­ Efforts to assess current urban forests at the national scale have est field data. Local city assessments provide urban forest data several limitations due to the gaps in urban forest data. To that are useful at the local scale, but local-scale assessments, in produce national or regional estimates of urban forests and their

19 The i-Tree Eco model is a new iteration of the UFORE (Urban Forest Effects) model and is one of several urban forest tools found within the i-Tree modeling suite. i-Tree is developed, maintained, and supported by a consortium of partners, including the Forest Service, Davey Tree, National Arbor Day Foundation, Society of Municipal Arborists, International Society of Arboriculture, and Casey Trees. Information on the model and other i-Tree applications is available at http://www.itreetools.org.

46 Assessing the Sustainability of Agricultural and Urban Forests in the United States effects, national or regional tree-cover data are combined with in Criterion 5 can likewise be addressed in this fashion and averages from various local urban forest assessments, and these could be improved with the addition of soil sampling and soils localities may not truly represent the overall region of analysis. information. Most of this type of information, however, will National and regional averages, from limited amounts and limited be the result of one-time analyses, which may provide useful spatial distribution of urban forest field samples, present chal­ information but will not result in the consistency across time lenges in providing truly accurate regional or national estimates. and space that is the ultimate goal of the MP C&I and similar reporting efforts. For that, a more comprehensive data gather­ Another challenge faced by national and regional urban forest ing and reporting effort combining remote-sensing capabilities estimates is the accuracy of the tree-cover data at these scales. with on-the-ground inventory sampling will be needed. Tree-cover maps based on 30-meter resolution images from the early 2000s are limited in their ability to accurately estimate tree As it currently stands, the nationally consistent data we have on cover, particularly in urban areas. Recent photo-interpretations urban forests enable us to partially address the following MP of tree cover nationally demonstrate that these tree-cover criteria: maps underestimate tree cover, on average, by 9.7 percent � Criterion 1: Conservation of biological diversity. Tree- with underestimation varying across the conterminous United cover data only, giving us an idea of the extent of forests but States (Greenfield et al. 2009; Nowak and Greenfield 2010). not their species structure or diversity. Fragmentation may Improved tree-cover estimates will provide a basis for better be measured and described using available data, but analysis estimates of urban forest structure, functions, and values from techniques will have to be developed. Tree counts can be local to broader scales, but these estimates will still be limited extrapolated from existing data (although these counts are by the absence of information from field data assessments (e.g., not considered in the MP C&I). number of trees, species composition, diameter distribution, tree health). This field derived information on urban forests is � Criterion 2: Maintenance of productive capacity. Rough lacking for most cities and regions of the United States. Most estimates of standing volume and volume growth can be data from field assessments are derived from individual efforts derived from forest cover information, but on-the-ground on the part of municipalities or regional collaborations. As a sampling is needed for greater precision. MP indicators on result, although the quantity and quality of urban forest data are timber and wood fiber production are not very applicable increasing, these data are not sufficient to adequately monitor in this context, but other output measures specific to urban or assess urban forests at the regional or national scale. forests may be devised.

Data adequacy relative to the Montréal Process Criteria � Criterion 5: Maintenance of forest contribution to global and Indicators (MP C&I) for Forest Sustainability. As carbon cycles. Carbon stocks and net sequestration on urban described previously, the information we currently have about forests can be estimated using forest cover and volume urban forest resources at the national level is largely restricted stocking estimates developed for Criteria 1 and 2. On-the- to urban forest cover estimates derived from the analyses of ground sampling is needed for greater precision, and overall satellite images or photo-interpretation, along with some esti­ carbon estimates for urban areas could be improved with the mates of tree counts and economic values modeled or otherwise addition of soil sampling. derived from the cover estimates and various field data assess­ ments. This information is much less than the database that The other MP criteria and many of the indicators in the three was assembled to address the 54 MP indicators covered in the criteria listed above currently cannot be adequately addressed National Report on Sustainable Forests—2010, hereafter the at the national level with available data. Although a number of National Report (USDA Forest Service 2011). Nonetheless, the these indicators are not very applicable in the realm of urban data on urban forests that we currently do have will go a long forestry, others, such as those covering forest health, are es­ way in addressing some of the key indicators on forest extent sential to understanding and managing urban forests. found in MP Criterion 1, and these data constitute an important foundation for developing a more comprehensive inventory. Knowledge of local or regional tree species distributions can Strategies for Improving Urban be cross-referenced with cover data to develop estimates of Forest Data total regional or national species counts and thereby potential susceptibility to pest epidemics and other pathogens. This Improvement in urban forest data gathering and reporting example is only one illustration of how the forest cover data activities can be accomplished by (1) synthesizing existing data can be used in conjunction with other data to address MP indi­ and standardizing data collection and formatting and (2) gather­ cators or other concerns at different spatial scales. Numerous ing more data from local to national scales. other possibilities exist. The MP indicators on carbon balances

Assessing the Sustainability of Agricultural and Urban Forests in the United States 47 Standardizing and Synthesizing this approach is the most likely strategy for addressing the data Available Data needs of the MP C&I. Moreover, even with a fully implemented inventory, this kind of synthetic approach will be essential As increasing amounts of urban tree and forest data become for addressing many of the social, economic, and institutional available, the ability to synthesize and report these data in a indicators that are found in Criteria 6 and 7 of the MP C&I. fashion useful to managers, planners, or policymakers working at local, regional, and national scales becomes paramount. Data New Data Gathering Opportunities collection efforts are currently not systematic, but rather op­ portunistic, being based on local managers’ desires and efforts and Challenges to collect and analyze urban forest data. A key challenge of this The most obvious means for attaining long-term and consistent effort is developing ways by which these local efforts can con­ data for urban forest analysis from the regional to national tribute to, and benefit from, data collection efforts elsewhere. scale is to integrate urban tree data collection within existing This integration can best be accomplished by developing forest inventory work under the FIA program, which currently consistent data collection protocols that can be used in different collects data for conventional forests across the entire United settings and then by providing consistent data reporting and States. In preparation for a national urban inventory, pilot analysis tools so that information can be easily shared. The testing of FIA plots and data collection techniques in urban i-Tree model, discussed previously, is an example of how this areas (Cumming et al. 2008) has been conducted in Indiana type of consolidation can be achieved on a voluntary basis. By (Nowak et al. 2007), Wisconsin (Cumming et al. 2007), New providing local practitioners with tools that make their jobs Jersey, Tennessee (Nowak et al. 2012), Alaska, California, easier, i-Tree facilitates consistent data generation and reporting, Colorado, Hawaii, Oregon, and Washington. A more recent allowing for comparability and (to a limited extent) aggregation action, the 2014 Farm Bill (formally the Agricultural Act of across space and time. Local reports are produced (e.g., Nowak 2014, Public Law 113-79), laid the legislative foundation for a et al. 2006b) and data can be combined with aerial cover analyses national urban inventory by directing FIA to revise its strategic to estimate regional- to national-scale characteristics of urban plan and describe the “organization, procedures, and funding forests. Although this activity does not take the place of an needed” to implement an annualized urban forest inventory. integrated national inventory, it does provide a wealth of data To this end, FIA has implemented a monitoring program that for understanding local conditions and for developing studies focuses on metropolitan regions and began data collection at broader scales. It can also provide an important source of in the Baltimore, MD, and Austin, TX, metropolitan areas in information for validating and augmenting broader inventory 2014 and additional areas in 2015 (i.e., Houston, TX; Madison, efforts. Regarding data standardization, efforts are also currently WI; Milwaukee, WI; Providence, RI; St. Louis, MO, and Des underway to develop international urban forest data-collection Moines, IA). FIA intends to sample and monitor more metro­ standards. politan areas in the coming years. Through the inclusion of additional metropolitan areas, a better national picture of urban In addition to the information developed by i-Tree, a great deal areas can be obtained over time. of disparate information is available from a wide variety of sources, ranging from municipal reports to academic studies A central question in institutionalizing a national inventory of and broader natural resource sampling efforts, such as the yearly urban forests is exactly what variables to measure. The i-Tree North American Breeding Bird Survey. With these various Eco urban variables have been developed and tested within the sources of data, the challenge becomes how to combine informa­ State urban pilot projects (Cumming et al. 2008) and provide a tion to better understand urban forests in a broader spatial and starting point for considering this question (see box 1). i-Tree social context. In most cases, this kind of work takes the form Eco is designed to be consistent with many standard FIA of one-time analyses that can contribute background information variables while simultaneously being responsive to the needs of supporting the type of consistent and repeated data-reporting urban foresters, and both professionals and volunteers can use efforts that are called for by the MP C&I. For those cases in it. A nationally instituted inventory of urban forests would dif­ which data collection efforts are ongoing, it may be possible fer somewhat from a typical i-Tree Eco local analysis in terms to institute analysis protocols to develop measures that can of variables and protocols for measurement, but the general be reported consistently across space and time. This type of analyses involving plot-level, tree-level, and environmental analysis, combining available data in an opportunistic fashion variables would be consistent. The new urban FIA monitoring from multiple sources, is, in fact, a key strategy in addressing program is integrating data collection with i-Tree variables so a number of the MP indicators on conventional forests in the that both i-Tree and FIA analysis programs can analyze the National Report, but it requires a sustained effort and an explicit data. Should this inventory be expanded to agricultural forests, commitment to consistency, which is not easy. Nonetheless, the degree to which these protocols would be adjusted to allow until a national urban forest inventory is fully implemented, for consistency across agricultural and urban forests will need

48 Assessing the Sustainability of Agricultural and Urban Forests in the United States Box 1. i-Tree Eco Sampling Variables* Plot-Level Information Tree Information (for individual tree measurements) Tree cover: The amount of the plot covered by tree canopy (in Land use (specific to tree) percent) Species Shrub cover: The amount of the plot covered by shrub canopy Status (records presence or removal of tree relative to past (in percent) inventory) Plantable space: Estimate of the amount of the plot area that Tree Characteristics is plantable for trees (in percent) � Total tree height Land Use � Height to live top � Actual land use(s): Required (e.g., residential, golf course, � Height to crown base park, commercial) � Crown width � Percent of area in each land use � Percent canopy missing Ground Cover � Crown dieback � Ground cover types present (e.g., bare soil, cement, grass) � Crown light exposure � Percent of area under each ground cover type � Percent impervious surface under the tree � Percent shrub cover under the tree Shrub Information � DBH (diameter at breast height) � Shrub species: Identify the shrub species � Shrub height Direction to building � Percent of total shrubs area Shortest distance to building � Percent of the shrub mass that is missing

*Abbreviated version. For more detail on included variables and sampling protocols see: www.itreetools.org.

to be addressed, as will be the potential inclusion of additional Cumming, A.B.; Twardus, D.B.; Nowak, D.J. 2008. Urban variables (e.g., soils) targeted at specific MP indicators or forest health monitoring: large scale assessments in the United related information needs. States. Arboriculture & Urban Forestry. 34(6): 341–346.

Donovan, G.H.; Butry, D. 2008. Market-based approaches to References tree valuation. Arborist News. August: 52–55. http://www. isa-arbor.com. (August 2015). Anderson, L.M.; Cordell, H.K. 1988. Influence of trees on Dozier, H. 2012. Invasive plants and the restoration of the residential property values in Athens, Georgia (USA): a survey urban forest ecosystem (Revised). In: Duryea, M.L.; Binelli, based on actual sales prices. Landscape and Urban Planning. E.K.; Korhnak, L.V., eds. Restoring the urban forest ecosystem. 15: 153–164. Circular 1266, Fact Sheet FOR98. [CD-ROM]. Gainesville, Anderson, L.M.; Mulligan, B.E.; Goodman, L.S. 1984. Effects FL: University of Florida, School of Forest Resources and of vegetation on human response to sound. Journal of Arbori­ Conservation, Florida Cooperative Extension Service, Institute culture. 10(2): 45–49. of Food and Agricultural Sciences. Chapter 9. http://www.edis. ifas.ufl.edu/pdffiles/fr/fr07300.pdf. (August 2015). Cook, D.I. 1978. Trees, solid barriers, and combinations: alternatives for noise control. In: Hopkins, G., ed. Proceedings: Dwyer, J.F. 1991. Economic value of urban trees. White paper national urban forest conference. Syracuse: SUNY, The State prepared in support of: A national research agenda for urban University of New York, College of Environmental Science forestry in the 1990s. Urbana, IL: International Society of and Forestry: 330–339. Arboriculture: 27–32.

Corrill, M.; Lillydahl, J.; Single, L. 1978. The effects of green­ Dwyer, J.F.; McPherson, E.G.; Schroeder, H.W.; Rowntree, belts on residential property values: some findings on the politi­ R.A. 1992. Assessing the benefits and costs of the urban forest. cal economy of open space. Land Economics. 54: 207–217. Journal of Arboriculture. 18(5): 227–234.

Cumming, A.B.; Nowak, D.J.; Twardus, D.B.; Hoehn, R.; Dwyer, J.F.; Schroeder, H.W.; Gobster, P.H. 1991. The signifi­ Mielke, M.; Rideout, R. 2007. Urban forests of Wisconsin cance of urban trees and forests: toward a deeper understanding 2002: pilot monitoring project 2002. State and Private Forestry of values. Journal of Arboriculture. 17: 276–284. Report NA-FR-05-07. Newton Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Area. 33 p.

Assessing the Sustainability of Agricultural and Urban Forests in the United States 49 Dwyer, J.F.; Schroeder, H.W.; Louviere, J.J.; Anderson, D.H. Nowak, D.J. 1993. Historical vegetation change in Oakland 1989. Urbanities willingness to pay for trees and forests in and its implications for urban forest management. Journal of recreation areas. Journal of Arboriculture. 15(10): 247–252. Arboriculture. 19(5): 313–319.

Greenberg, C.H.; McNab, W.H. 1998. Forest disturbance in Nowak, D.J. 1994. Air pollution removal by Chicago’s urban hurricane-related downbursts in the Appalachian Mountains forest. In: McPherson, E.G.; Nowak, D.J.; Rowntree, R.A., eds. in North Carolina. Forest Ecology and Management. 104: Chicago’s urban forest ecosystem: results of the Chicago urban 179–191. forest climate project. Gen. Tech. Rep. NE-186. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Greenfield, E.J.; Nowak, D.J.; Walton, J.T. 2009. Assessment Northeastern Research Station: 63–81. of NLCD percent tree and impervious surface cover estimates. Photogrammetric Engineering and Remote Sensing. 75(11): Nowak, D.J. 2000. The interactions between urban forests and 1279–1286. global climate change. In: Abdollahi, K.K.; Ning, Z.H.; Ap­ peaning, A., eds. Global climate change and the urban forest. Heisler, G.M. 1986. Energy savings with trees. Journal of Baton Rouge: Gulf Coast Climate Change Assessment Council Arboriculture. 12(5): 113–125. (GCRCC) and Franklin Press: 31–44. Interagency Working Group on Social Cost of Carbon, United Nowak, D.J. 2010. Urban biodiversity and climate change. In: States Government (Interagency Working Group). 2015. Social Muller, N.; Werner, P.; Kelcey, J.G., eds. Urban biodiversity cost of carbon for regulatory impact analysis. Technical Sup­ and design. Hoboken, NJ: Wiley-Blackwell Publishing: 101–117. port Document. Under Executive Order 12866. https://www. whitehouse.gov/sites/default/files/omb/inforeg/scc-tsd-final- Nowak, D.J. 2012. Contrasting natural regeneration and tree july-2015.pdf. (August 2015). planting in 14 North American cities. Urban Forestry and Urban Greening. 11: 374–382. Intergovernmental Panel on Climate Change (IPCC). 2007. Climate change 2007: the physical science basis—summary for Nowak, D.J.; Crane, D.E.; Dwyer, J.F. 2002a. Compensatory policymakers. Geneva, Switzerland: Intergovernmental Panel value of urban trees in the United States. Journal of Arboricul­ on Climate Change Secretariat. 18 p. http://www.ipcc.ch/pdf/ ture. 28(4): 194–199. assessment-report/ar4/wg1/ar4-wg1-spm.pdf. (August 2015). Nowak, D.J.; Crane, D.E.; Stevens, J.C. 2006a. Air pollution Irland, L. 2000. Ice storms and forest impacts. Science of the removal by urban trees and shrubs in the United States. Urban Total Environment. 262: 231–242. Forestry and Urban Greening. 4: 115–123.

Iverson, L.R.; Prasad, A.M. 2001. Potential changes in tree Nowak, D.J.; Cumming, A.; Twardus, D.; Hoehn, R.E.; Brandeis, species richness and forest community types following climate T.J.; Oswalt, C.M. 2012. Urban forests of Tennessee, 2009. change. Ecosystems. 4: 186–199. Gen. Tech. Rep. SRS-149. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 52 p. Johnston, M. 2004. Impacts and adaptation for climate change in urban forests. Paper presented at the 6th Canadian urban Nowak, D.J.; Dwyer, J.F. 2007. Understanding the benefits forest conference. 15 p. and costs of urban forest ecosystems. In: Kuser, J., ed. Urban and community forestry in the Northeast. New York: Springer: Kuo, F.E.; Sullivan, W.C. 2001a. Environment and crime in 25–46. the inner city: Does vegetation reduce crime? Environment and Behavior. 33(3): 343–365. Nowak, D.J.; Greenfield, E.J. 2010. Evaluating the National Land Cover Database tree canopy and impervious cover Kuo, F.E.; Sullivan, W.C. 2001b. Aggression and violence estimates across the conterminous United States: a comparison in the inner city: impacts of environment via mental fatigue. with photo-interpreted estimates. Environmental Management. Environment and Behavior. 33(4): 543–571. 46: 378–390. Liebhold, A.M.; MacDonald, W.L.; Bergdahl, D.; Mastro, V.C. Nowak, D.J.; Greenfield, E.J. 2012a. Tree and impervious 1995. Invasion by exotic forest pests: a threat to forest ecosys­ cover in the United States. Landscape and Urban Planning. tems. Forest Science, Monograph 30: a0001–z0001(1). 107: 21–30. National Acid Precipitation Assessment Program (NAPAP). Nowak, D.J.; Greenfield, E.J. 2012b. Tree and impervious 1991. 1990 integrated assessment report. Washington, DC: cover change in U.S. cities. Urban Forestry and Urban Green­ National Acid Precipitation Assessment Program. 520 p. ing. 11: 21–30.

50 Assessing the Sustainability of Agricultural and Urban Forests in the United States Nowak, D.J; Greenfield, E.J.; Hoehn, R.; LaPoint, E. 2013. Pye, J.M. 1988. Impact of ozone on the growth and yield of Carbon storage and sequestration by trees in urban and com­ trees: a review. Journal of Environmental Quality. 17: 347–360. munity areas of the United States. Environmental Pollution. Reethof, G.; McDaniel, O.H. 1978. Acoustics and the urban 178: 229–236. forest. In: Hopkins, G., ed. Proceedings of the National Urban Nowak, D.J.; Hirabayashi, S.; Ellis, E.; Greenfield, E.J. 2014. Forest Conference. Syracuse: SUNY, The State University of Tree and forest effects on air quality and human health in the New York, College of Environmental Science and Forestry: United States. Environmental Pollution. 193: 119–129. 321–329.

Nowak, D.J.; Hoehn, R.; Crane, D.E.; Stevens, J.C.; Walton, Sanders, R.A. 1986. Urban vegetation impacts on the urban J.T. 2006b. Assessing urban forest effects and values: Washing­ hydrology of Dayton, Ohio. Urban Ecology. 9: 361–376. ton, D.C.’s urban forest. Res. Bul. NRS-1. Newtown Square, Schroeder, H.W. 1989. Environment, behavior, and design PA: U.S. Department of Agriculture, Forest Service, Northern research on urban forests. In: Zube, E.H.; Moore, G.L., eds. Research Station. 24 p. Advances in environment, behavior, and design. New York: Nowak, D.J.; Hoehn, R.E.; Crane, D.E.; Stevens, J.C.; Walton, Plenum Press: 87–107. J.T.; Bond, J. 2008. A ground-based method of assessing urban Schroeder, H.W.; Anderson, L.M. 1984. Perception of personal forest structure and ecosystem services. Arboriculture and safety in urban recreation sites. Journal of Leisure Research. Urban Forestry. 34(6): 347–358. 16: 178–194. Nowak, D.J.; Noble, M.H.; Sisinni, S.M.; Dwyer, J.F. 2001. Sommer, R.; Learey, F.; Summit, J.; Tirrell, M. 1994a. Social Assessing the U.S. urban forest resource. Journal of Forestry. benefits of resident involvement in tree planting: compressions 99(3): 37–42. with developer-planted trees. Journal of Arboriculture. 20(6): Nowak, D.J.; Stein, S.M.; Randler, P.B.; Greenfield, E.J.; 323–328. Comas, S.J.; Carr, M.A.; Alig, R.J. 2010. Sustaining America’s Sommer, R.; Learey, F.; Summitt, J.; Tirrell, M. 1994b. Social urban trees and forests. Res. Bul. GTR-NRS-62. Newtown benefits of residential involvement in tree planting. Journal of Square, PA: U.S. Department of Agriculture, Forest Service, Arboriculture. 20(3): 170–175. Northern Research Station. 28 p. Spyratos, V.; Bourgeron, P.S.; Ghil, M. 2007. Development Nowak, D.J.; Stevens, J.C.; Sisinni, S.M.; Luley, C.J. 2002b. at the wildland-urban interface and the mitigation of forest- Effects of urban tree management and species selection on fire risk. Proceedings of the National Academy of Sciences. atmospheric carbon dioxide. Journal of Arboriculture. 28(3): 104(36): 14272–14276. 113–122. Stolte, K. 1996. The symptomology of ozone injury to pine Nowak, D.J.; Twardus, D.; Hoehn, R.; Mielke, M.; Smith, B.; foliage. In: Miller, P.R.; Stolte, K.W.; Duriscoe, D.M.; Pronos, Walton, J.; Crane, D.; Cumming, A.; Lake, M.; Marshall, P. J., tech. coords. Evaluating ozone air pollution effects on pines 2007. National Forest Health Monitoring Program, monitoring in the Western United States. Gen. Tech. Rep. PSW-GTR-155. urban forests in Indiana: pilot study 2002, part 2: statewide Albany, CA: U.S. Department of Agriculture, Forest Service, estimates using the UFORE model. Northeastern Area Report. Pacific Southwest Research Station: 11–18. NA-FR-01-07. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 13 p. Syphard, A.D.; Radeloff, V.C.; Keeley, J.E.; Hawbaker, T.J.; Clayton, M.K.; Stewart, S.I.; Hammer, R.B. 2007. Human Nowak, D.J.; Walton, J.T. 2005. Projected urban growth and influence on California fire regimes. Ecological Applications. its estimated impact on the U.S. forest resource (2000–2050). 17(5): 1388–1402. Journal of Forestry. 103(8): 383–389. Taylor, A.F.; Kuo, F.E.; Sullivan, W.C. 2001a. Coping with Nowak, D.J.; Walton, J.T.; Dwyer, J.F.; Kaya, L.G.; Myeong, ADD: the surprising connection to green play settings. Envi­ S. 2005. The increasing influence of urban environments on ronment and Behavior. 33(1): 54–77. U.S. forest management. Journal of Forestry. 103(8): 377–382. Taylor, A.F.; Kuo, F.E.; Sullivan, W.C. 2001b. Views of nature Pouyat, R.V.; Yesilonis, I.D.; Nowak, D. 2006. Carbon storage and self-discipline: evidence from inner-city children. Journal by urban soils in the United States. Journal of Environmental of Environmental Psychology. 21: 49–63. Quality. 35: 1566–1575. Ulrich, R.S. 1984. View through a window may influence Proulx, O.J.; Greene, D.F. 2001. The relationship between recovery from surgery. Science. 224: 420–421. ice thickness and northern hardwood tree damage during ice storms. Canadian Journal of Forest Research. 31: 1758–1767.

Assessing the Sustainability of Agricultural and Urban Forests in the United States 51 U.S. Department of Agriculture (USDA), Forest Service. 2011. Westphal, L.M. 1999. Empowering people through urban National report on sustainable forests—2010. FS-979. Wash­ greening projects: Does it happen? In: Kollin, C., ed. Proceed­ ington, DC: U.S. Department of Agriculture, Forest Service. ings: 1999 national urban forest conference. Washington, DC: 212 p. http://www.fs.fed.us/research/sustain/. (August 2015). American Forests: 60–63.

U.S. Department of Commerce, Census Bureau (Census Westphal, L.M. 2003. Urban greening and social benefits: a Bureau). 2014. Geography. https://www.census.gov/geo/ study of empowerment outcomes. Journal of Arboriculture. reference/ua/urban-rural-2010.html. (August 2015). 29(3): 137–147.

U.S. Environmental Protection Agency (EPA). 2009. Climate Wolf, K.L. 2003. Public response to the urban forest in inner-city change. http://www.epa.gov/climatechange/. (August 2015). business districts. Journal of Arboriculture. 29(3): 117–126.

U.S. Environmental Protection Agency (EPA). 2015. The Wolf, K.L. 2004. Trees and business district preferences: a case social cost of carbon. http://www.epa.gov/climatechange/ study of Athens, Georgia US. Journal of Arboriculture. 30(6): EPAactivities/economics/scc.html. (August 2015). 336–346.

U.S. Geological Survey (USGS). 2008. Multi-resolution land Zhang, Y.; Zhang, D.; Schelhas, J. 2005. Small-scale nonin­ characteristics consortium. http://www.mrlc.gov. (August 2015). dustrial private forest ownership in the United States: rationale and implications for forest management. Silva Fennica. 39(3): Valinger, E.; Fridman, J. 1997. Modeling probability of snow 443–454. and wind damage in Scots pine stands using tree characteristics. Forest Ecology and Management. 97: 215–222.

Webb, S.L.; Pendergast, T.H.; Dwyer, M.E. 2001. Response of native and exotic maple seedling banks to removal of exotic, invasive, Norway maple. Journal of the Torrey Botanical Society. 128: 141–149.

52 Assessing the Sustainability of Agricultural and Urban Forests in the United States